Cambridge Azur 851A integrated amplifier Measurements

Sidebar 3: Measurements

I measured the second sample of the Cambridge Audio Azur 851A (serial no. FE C10427K 1205 0137) using my recently recalibrated Audio Precision SYS2722 system (see, and the January 2008 "As We See It"). Before performing any measurements on an amplifier, I run it for an hour at one-third its specified maximum power, which, thermally, is the worst case for an amplifier with a class-B or -AB output stage. I cut short this preconditioning with the Cambridge 851A after 30 minutes, however, because its case was getting very hot—the top-panel grille over the internal heatsinks was too hot to touch, at 148.4°F (64.7°C)—and I was concerned that something might break. I don't believe this will be a problem in normal use, as long as the amplifier is given plenty of space for ventilation.

As set-up, the maximum voltage gain into 8 ohms for the unbalanced inputs was 38.9dB, which is typical for an integrated amplifier. For balanced inputs, however, the maximum gain was exactly 6dB lower. All inputs preserved absolute polarity; ie, were non-inverting. The input impedance was close to specification, at 19.5k ohms across the audioband for the unbalanced inputs and for each phase of the balanced inputs.

The output impedance was very low, at 0.07 ohm (including 6' of speaker cable) in the bass and midrange, rising to 0.1 ohm at the top of the audioband. With this low impedance, there is just ±0.1dB of variation in the amplifier's frequency response when driving our standard simulated loudspeaker (fig.1, gray trace). This graph was taken with the amplifier in Direct mode, bypassing the tone controls, and with the volume control set to its maximum. Reducing the volume control to an indicated –12dB on the front-panel display didn't change the Azur 851A's response, and the amplifier has a very wide small-signal bandwidth into 8 ohms (blue and red traces). The response is –1.75dB at the 200kHz limit of this graph, and even into 2 ohms (fig.1, green trace), the response is down by less than 0.2dB at the top of the audioband. This gives a 10kHz squarewave very short risetimes (fig.2), and, commendably, there is no trace of overshoot or ringing. The 851A's reproduction of a 1kHz squarewave (fig.3) was essentially perfect.

Fig.1 Cambridge Azur 851A, volume control set to maximum, Direct mode, frequency response at 2.83V into: simulated loudspeaker load (gray), 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta), 2 ohms (green) (0.5dB/vertical div.).

Fig.2 Cambridge Azur 851A, small-signal, 10kHz squarewave into 8 ohms.

Fig.3 Cambridge Azur 851A, small-signal, 1kHz squarewave into 8 ohms.

Activating the Treble and Bass tone controls (I successfully unstuck the Treble control knob with a toothpick) and setting them to their maximum and minimum positions gave the traces shown in fig.4. These controls are shelving types that produce up to ±10dB measured change in response below 30Hz and above 40kHz. The change in response at 80Hz and 10kHz was a more modest and appropriate ±6dB. Channel separation (not shown) was 78dB in both directions below 5kHz. The wideband, unweighted signal/noise ratio (ref. 2.83V into 8 ohms with the volume set to its maximum) was modest at 57.8dB, but improved to 74.8dB when the measurement bandwidth was reduced to the audioband, and 77.2dB when A-weighted. Spectral analysis of the amplifier's output while it drove a 1kHz tone at 1W into 8 ohms (fig.5) indicated that the supply-related component at 120Hz was the highest-level noise component, at –86dB, though other supply-related components are also present. Experimenting with the grounding between the analyzer and the amplifier didn't change this behavior.

Fig.4 Cambridge Azur 851A, response with Bass and Treble controls set to "0," maximum, minimum (left channel blue, right red; 2dB/vertical div.).

Fig.5 Cambridge Azur 851A, spectrum of 1kHz sinewave, DC–1kHz, at 1W into 8 ohms (linear frequency scale).

Fig.6 plots the left channel's percentage of THD+noise against output power into 8 ohms. The downward slope of the trace below 100W reveals that the measurement is dominated by noise, actual distortion components not rising above the noise floor until the amplifier is close to clipping. However, it wasn't possible to measure the Cambridge's power at our usual definition of clipping, 1% THD+N, because at 136.4Wpc into 8 ohms (21.35dBW, 0.55dB above the specified maximum power of 120W), the 851A's protection circuit operated the volume control to reduce the level, as indicated by the downward step in this graph above 136.4Wpc. The same thing happened with both channels driven into 4 ohms (fig.7), with the volume control now being taken over above 220.5Wpc (20.4dBW).

Fig.6 Cambridge Azur 851A, left channel, distortion (%) vs 1kHz continuous output power into 8 ohms.

Fig.7 Cambridge Azur 851A, left channel, distortion (%) vs 1kHz continuous output power into 4 ohms.

Looking at the distortion waveform in the left channel at a reasonably high power, 40W into 4 ohms, there is just a hint of second harmonic visible (fig.8, bottom trace), even though, to lower the contribution of noise in this graph, I averaged 64 captures. However, when I plotted the THD+N percentage against frequency at 20V into 8 and 4 ohms (fig.9), respectively equivalent to 50 and 100Wpc, while both channels offered the expected low distortion into 8 ohms (blue and red traces), the right channel offered higher distortion into 4 ohms (magenta). I repeated the clipping test into 8 ohms with the right channel; when this result (fig.10) is compared with fig.5, it can be seen that the distortion in the right channel begins to rise above the noise floor above a few tens of watts, though it remains below 0.04% below 100W.

Fig.8 Cambridge Azur 851A, left channel, 1kHz waveform at 40W into 4 ohms, 0.007% THD+N (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.9 Cambridge Azur 851A, THD+N (%) vs frequency at 20V into: 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta).

Fig.10 Cambridge Azur 851A, right channel, distortion (%) vs 1kHz continuous output power into 8 ohms.

Fig.11, taken at a relatively high power into 4 ohms, reveals that the right channel (red trace) has a much higher proportion of both odd- and even-order distortion products than does the left (blue), with the third harmonic the highest in level at –69dB (0.03%). I don't believe this will give rise to audible problems, but this behavior does suggest that Cambridge's "Class XD (crossover displacement)" circuit was not working correctly in the 851A's right channel. Even so, when I tested for intermodulation distortion at a very high power into 4 ohms, I found no distortion products higher than –80dB (0.01%) (fig.12).

Fig.11 Cambridge Azur 851A, spectrum of 50Hz sinewave, DC–1kHz, at 80W into 4 ohms (linear frequency scale).

Fig.12 Cambridge Azur 851A, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 150W peak into 4 ohms (linear frequency scale).

Assuming that the less-good performance of its right channel was a sample-specific fault, Cambridge Audio's Azur 851A is a well-built amplifier that offers a lot of power with very low distortion at an affordable price.—John Atkinson

Cambridge Audio
US distributor: Audio Plus Services
156 Lawrence Paquette Industrial Drive
Champlain, NY 12919
(800) 663-9352

iosiP's picture

I mean, the first sample did not work, the second had problems in the right channel. And these are samples send to reviewers, not purchased by the unsuspecting customer.
I know they are making them in China but for close to $2k I certainly expect better QC!

Erik's picture

I've been wanting to demo this amp for a while and this didn't change that one bit!

Totally get what you say about texture and color, and from one junkie to another, you should have a look at the Hegel H80. Never mind the built in DAC, the amp gels superbly with the Rega in apodizing mode.

BaTou069's picture

Hello Herb,
Great Review! You mentioned the Rogue Audio Sphinx as Associated Equipment but didn't write a word of comparison. I find this comparison rather interesting, both integrated being fully analog, almost same price range, but different approach. Could you fill in? Thanks's picture

One blew up, one had other major problems. Makes me want to stay away from the whole company. Consumers dont have test gear, we sort of have to hope that things work..... Can you imagine what a hassle the dealer would give you if you tried to return a product saying the right channel sounds distorted? Its great that stereophile can find these faults with measurements, but that does little for the average consumer who just got screwed. Terrible. All cambridge product are off my list

damir's picture

My Cambridge Audio 840C cd player died twice.

First fix cost me 700 USD.

A year later, it failed, the same way it failed the first time!

I gave up on Cambridge Audio, although that DAC was really doing magic while it worked.

Same thing I can say about NAD C370... had two of them bridged... both died the same way.

Right now, I'm on NAD m51 and m3... fingers crossed master series will last longer.

corrective_unconscious's picture

What problem was that exactly that happened twice to the CD player? It was out of warranty and you repaired a $1200 to $1500 component for $700.... What shop or dealer handled the $700 repair? I assume you contacted the same people when it happened the second time...whether out of the repair warranty period or not.

damir's picture

Yes, it was just out of warranty, I paid for the fix. I liked the sound and several digital inputs it offered.

There is a loud noise coming from the left channel when you first turn on the CD player. Turn off, turn on, turn off... let it warm up and after some time the *loud* white noise stops and you can start using it.

They replaced the complete motherboard the first time.

A year later, it happened again. I am not going to spend 700 USD again to fix it.

Cambridge Audio - never again.

corrective_unconscious's picture

It was about a $1200 component, at least two years old, which means depreciated to about $600 or $700, and you spent $700 to get it fixed.

There's a first time for everything. I wouldn't use the phrase "motherboard" for a CD player, personally. What board? A power supply board? A digital chip board? What were the symptoms of this failure? It made funny noises but after warmup you could use it normally?

What dealer did the repair? What did they say when the exact same "motherboard" problem re occurred? Didn't you complain to the dealer and to the manufacturer after the expensive repair failed in the same way? What did they have to say for themselves?

I found a nice image of the guts (C rev.) for us to look at:

mendaily DOT com/wp-content/uploads/2012/12 FORWARD SLASH Cambridge-Audio-Azur-851C-Inside DOT jpg

damir's picture

I don't understand what is confusing you. I did not want to buy a new one, I fixed the old one.

They changed the complete *motherboard* with DAC, CPU's, condensators, external connectors, everything.

What dealer? The one that represents cambridge audio and sells the CD player.

Why would I complain again? What could they say? It lasted for a bit over a year and died just like it did the first time.

I wrote in the previous message, it had loud white noise going on a left channel, until it warmed up.

Here is the photo of the motherboard:

FlyhiG's picture

Your report mirrors my experience with CA products. Great sound nice features but not reliable aa one would expect. Sold by the better stores, so not inexpensive. Just too many other makers for me to take a chance on CA gear again.

mrd745_2000's picture

In many stereophile articles you mention where the product is made. I suggest you mention it for every product. I have had bad luck with made in china products. The manufacturers seem to think this is acceptable since they provide us with a short warranty period. I disagree. This gear cost a lot of money. We should get a reliable product that should last for years. But I have made my choice. If I cannot afford a north american or european product I will buy something used or wait. Manufacturers comments that they monitor quality contol or own the factory are meaningless. No more made in China for me.

Sugarbear66's picture

OK I got mine in 2012 and LOVED it, by 2014 it was randomly turning off and weird volume knob issues. (turning it down turned it up) so I had to use remote but, that was fine. But, the turning off got worse and so 6 weeks of a repair later it came back working great and now it's 2016 I'm sending it back again for repair. I've finally lost patience with it, I love the way it sounds, I love the aesthetics of it, it's a 2k integrated but, I had to break up with it. I Just installed a Rogue Audio Sphinx v2 in my rig to replace it during the 851 repair and it's a dream. Maybe I got a lemon 851a. But, no matter my love for the Cambridge it the quirkiness did me in in the end. Happy with my Rogue.

allhifi's picture

SB66: Good call.

I have a 840E preamp -great piece, but damn the under-spec'd resistor/relay 'volume attenuation' (that constantly overheats -at high attenuation- causing hi-fi buff's to fear losing their prized kit. i.e. the up/down/runaway volume).

I understand completely the frustration, however, I've not given up on my beloved 840E (pre-amp); believing I can find a suitable (much higher quality/power/heat-dissipation resistor-relay 'board' (or necessary parts).

To think that CA couldn't fix that what plagued the 8-400 series (6-10 years on -in the 8-500 series is both laughable -and sad. For, as you say, CA otherwise has shown it can make some impressive kit.

Go figure,


allhifi's picture

Hey Frems: Solid review. Cambridge Audio has some fine product; the 851A reviewed case in point.

A few years back I also stumbled, fortuitously it proved, on a CA product; 840E preamplifier. One heck of a pre -a rare (musical) treat.

Then, as now it appears, CA has both some impressive engineering/listening 'chops', but also some glaring faux pas; the 840E's resistor/relay volume attenuation used under-spec'd (power/heat) resistors tat result in 'runaway/ shorting' volume level -replete with accompanying pops, crackles ... hellish fear -of our beloved hifi (amps/speakers).

I, as you, agree:

" ... My only caveats concern its .... nice little plastic tone controls, one of which got stuck and stopped working ...hermetically sealed, military-spec pots and switches? If they work in an F16, they'll surely outlast all these fashionable, programmable, namby-pamby, chip-based interfaces."

I'd add; dispense with those ridiculous "tone" controls altogether. And return to using a higher-spec'd resistor/relay volume implementation.

If only CA could rid themselves of a few of these glaring mis-steps, they'd have a stunning, high-performance, high-value offering that would satisfy audiophile's (rightful) expectations.

peter jasz